Design of genetic fuzzy parallel parking control systems

Automatic parallel parking is an important capability for autonomous ground vehicles (AGVs) in both commercial and military applications. Fuzzy logic controllers for parallel parking control of skid steering AGVs were developed in previous research, where the membership functions and scaling factors of fuzzy systems were tuned manually by trial and error. The first aim of this study is to adjust the parameters of the membership functions and scaling factors for the previously developed fuzzy parallel parking algorithm by using a genetic algorithm. The GA implementation details, such as the design parameters and choice of fitness function, are described. Also, the previously developed fuzzy parallel parking algorithm for skid steering system is extended to AGVs with front-wheel steering. This study shows that the fuzzy algorithm is valid for front-wheel steering systems; only the membership functions and scaling factors must be modified.